Attempt to screen capture Fusion360 issue fails, component tree missing.

[i-make-robots]

- Windows 10, Fusion360 2.0.2865 running.

- Open project.

- Windows+G "This is a game" (screen recording begins)

- Surf over component tree.

- Bug I was trying to show is that hovering over some lightbulbs (show/hide) causes just that one component in the tree to redraw such that the expand arrow on the left disappears and the lightbulb moves to the left ~30 px.  Also sometimes happens when trying to expand the tree, instead I hide the component.  Expanding the tree causes the selected item to jump up the screen, instead of being well behaved and staying under my cursor.

- Windows+G "stop recording"

- Open video in Adobe Premiere.  Component tree missing.  Cursor flying around over nothing.  All other features (sidebar, model, timeline, etc) appear to be where they belong.

Ended up filming it with a gopro.

https://www.youtube.com/watch?v=vBE_LorUdSM

[Phil.E]

Hi,

This is a bit of a concern. We have fixed this before, so it looks like a regression, however before it was easy to achieve on any machine. I can't reproduce it at all on several machines.

First, since this does not reproduce on all machines. So we should find out more about your hardware.

• Updated video drivers lately?
• What is your graphics diagnostic information? (under the ?Help menu)
• How many monitors
• Laptop?
• 4K?

Second, is it data related? Can you at any time "not" achieve it, such as in a brand new or lightly populated design?

Regarding the video capture. Sounds like a problem with the way windows captures video. It's only recording the active window. The other items are separate windows. 

There are two free tools we use that never fail: Screencast by Autodesk, and JING by techsmith. Screencast is best because you can embed them here and they show the command usage on a graph. JING is a little more "point and shoot" and gets results very quickly. Either way, no need to film yourself working with a physical camera.

I'm going to open a bug on this, assuming it's not a regression of the previous problem because it's not exactly the same. If any other customers want to chime in here to help figure this out it would be appreciated. The key to fixing any bug is reproducing it on a debug machine.

Thanks for your help and thanks for the post.